Discussion Papers,
Statistics Norway, Research Department

Abstract: This paper introduces a novel multivariate composite estimator for the Labour Force Survey (LFS). Unlike the univariate composite estimators used in some countries, the multivariate estimator takes into account the different probabilities of transitioning between labour market categories, such as employment, unemployment, or non-participation in the labour force. By directly estimating all categories, it avoids residual determination issues, where one category is estimated as the difference between the total population and the sum of others. The multivariate approach improves the accuracy of the population estimates for each category. Additionally, the paper introduces a method to account for time-varying biases associated with how long the respondents have participated in the survey, explicitly incorporating wave-specific effects and their evolution over time.

Keywords: Autocorrelated sampling errors; Labour Force Survey (LFS); Multivariate composite estimator; Population estimation accuracy; Survey wave-specific biases; Transition probabilities in labour market categories

JEL-codes: C13; C83; J21

34 pages, April 2025

Full text files

DP1022.pdf  

Download statistics

• Statistics Norway, Research Department
• Home page for this series

Questions (including download problems) about the papers in this series should be directed to L Maasø ()
Report other problems with accessing this service to Sune Karlsson ().

RePEc:ssb:dispap:1022This page generated on 2026-01-07 04:36:24.